- A new study finds that humans are using words that are commonly found in Ia chatbots more frequently.
- The researchers found a measurable change towards formal language patterns commonly used by the LLM.
- The words that appear most frequently include meticulous, delve and underlined.
There is a lot of concern that people use human discourse and writing, but a new study of the Florida State University has found that there is a change in the other direction. It turns out that people use common words to chatbots of the most often in a casual conversation. Terms such as Garner, Delve, Intrince and Underscore are appearing as the last comedy sentence thanks to Chatgpt and his compatriots.
The study showed that AI is shaping how we talk, if not the issues itself. The way in which we form sentences, we choose words and try to be formal are affected since the launch of Chatgpt in 2022. Words AI the chatbots could in excess, their style of signature, it is becoming a basic element of human conversation, at least when talking about science and technology.
The researchers analyzed more than 22 million words of speech without science and technology podcasts, were based on an improvised discussion using an informal natural language. Chatgpt’s favorite words now prevail in discussions, much more than before his debut. To be fair, many of the words connote formal and technical value and are the basic vernacular in certain contexts. However, no one said meticulous, strategically, surpassed or boasted about these podcasts almost as much as they have been in the last three years.
In particular, the simplest synonyms are not emerging so much. The researchers considered much more likely to say underlined, do not accentuate, or deepen, but do not explore. It seems that people absorb specific stylistic tics of the text generated by AI. It is a bit like when a child begins to use a term that listens to someone who admits to use a lot, only in this case, they are adults who mimic a language model trained in billions of words. And it is not the only evidence that it is happening. A recent study found that the same thing among academics happened. T
Ai thessaurus
That may sound as a harmless side effect of the digital era, but there is a potential worrying consequence due to the possibility of a feedback circuit. The researchers suggest that AI could be accelerating linguistic evolution, but admit that it could actually change the way we talked in a way that would never have happened without the influence of the chatbots of AI.
This is not the first time that language has changed under technological pressure. Text messages introduced abbreviations and emojis. Social networks gave us hashtags, Gif reaction and a lot of inscrutable jargon for people who are not in their adolescence. But, this could be a bigger treatment, and not in good sense. The problem is that these phrases not only add to how we talk, but also replace more vocabulary of free and evocative wheels and make everyone feel like the same soft digital assistant.
It is not an Apocalypse language at this time in any way. To say something is great instead of convincing will not confuse most people. But the future of how we believe that we will be affected by how we talk, and the way in which AI expresses things could affect the way we frame the ideas and make decisions, in a way that we do not appreciate. Researchers will have to deepen more studies and see if the results obtained underline the initial responses.
